A radio frequency (RF) power amplifier is a key component of a network equipment (NE) in a radio communication system. The RF power amplifier is substantially an energy converter, which converts Direct Current (DC) energy of a power source into RF energy for transmission through an antenna. A ratio of the RF power to the DC power provided by the power source is referred to as the efficiency η of the power amplifier, and the efficiency is an important factor of the power amplifier. Taking the power amplifier in a base station for example, the efficiency is directly associated with factors such as the power source, heat dissipation, size, fans, and noises of the base station system. A high efficiency of the power amplifier improves the reliability of the base station system and reduces the cost of the base station equipment. For a telecommunication operator, a high efficiency of the power amplifier can effectively reduce the cost of system operation and subsequent maintenance.
The efficiency η of a power amplifier is calculated according to the following formulae:η=(Power_rf/power_dc)×100%  [1]where Power_rf is an RF power, and Power_dc is a DC power; andPower_dc=Vdd×Id  [2]where Vdd is a voltage provided by a DC power source, and Id is a current provided by the DC power source.
In the whole power conversion process, a part of the DC energy is inevitably converted into heat, which will be wasted. Therefore, the actual efficiency of the power amplifier is always lower than 100%.
In a current base station power amplifier, to consider the efficiency and linearity indexes comprehensively, a static working point of the power amplifier is normally set to Class A or Class AB, that is, a static working current of the power amplifier Idq>0 A Taking a laterally diffused metal oxide semiconductor (LDMOS) for example, the LDMOS is a power amplifier transistor widely applied at present. If a voltage bias of a power amplifier using the power amplifier transistor is in Class AB, and the power amplifier is in a saturated output state, the efficiency is the highest. However, with the decrease of the output power, the efficiency will be reduced gradually. That is, the ratio of the heat converted from the DC energy provided by the power source will rise with the decrease of the efficiency of the power amplifier.
When the power amplifier does not output any RF power, the dissipated DC power is as follows:Power_dc=Vdd×Idq  [3]
For the base station system, the state in which the power amplifier does not output any RF power appears frequently (for example, when no subscriber accesses the system). According to the above analysis, at this time, for a common Class A or Class AB amplifier, the static power is wasted, and the overall efficiency of the power amplifier is lowered.
To improve the overall efficiency of the power amplifier, the following solution is adopted in the prior art. When the power amplifier is in a state with no RF power output, a voltage of a drain electrode of the power amplifier transistor in the power amplifier is adjusted to 0 V. It is known from Formula [3] that, at this time, Power_dc=0 W i.e. the dissipated DC power is 0 W.
Though the overall efficiency of the power amplifier is increased to some extent by using the above method, the inventors found the following problems from the solution.
First, the response time is long, the solution is applicable to few scenarios only, and the improvement to the efficiency of the power amplifier is limited.
Normally, the drain electrode of the power amplifier transistor operates in the state of high voltage (for example, 28 V) and large current (for example, 10 A), so the power supply unit thereof must be a power source capable of providing a high power. Limited by factors such as the charging and discharging of high-capacitance capacitors and the soft-start mechanism to ensure security, the time for establishing or disabling the output voltage of such a power source is often several seconds or even several tens of seconds.
However, in normal situations, time periods required by services of the base station system are much shorter than one second. For example, in a Global System for Mobile Communications (GSM), the timeslot period of each user is only 577 μs. Subscribers may access (the power amplifier needs to switch on) or not access (the power amplifier needs to switch off) the GSM system in a timeslot period, while the solution of controlling the voltage of a drain electrode cannot track such a fast change. To ensure normal communications, the voltage on the drain port of the power amplifier transistor must remain at the normal operating voltage without changes for a long time, and thus a part of the static power will be dissipated
In view of the above, the solution of controlling the voltage of a drain electrode is applicable to very few scenarios in practice. Normally, this solution is adopted only when no subscriber accesses the system for a long time at night. Therefore, the improvement to the efficiency of the power amplifier is unobvious, and the power-saving effect is limited.
In addition, the control circuit is complicated with a high cost and low reliability.
The solution of controlling the voltage of a drain electrode deals with signals with a high voltage and large current, so many high-power elements are needed. Therefore, the circuit implementation is complicated, the cost is high, and the reliability is low, which may easily cause potential quality problems.